Abstract

Although well-maintained roads are an important source of transportation, their economic impact is rarely felt in a nation. Here, drivers who aren't convinced of their safety are seriously at risk from potholes and speed breakers. Potholes, poor night vision, inaccurate speed breaker signage, and the driver's carelessness all contributed to this disaster. All of these elements harm the vehicle's suspension system in addition to having an impact on drivers who are on the road. These days, cars with reduced ground clearance have trouble on uneven roads. In this paper, we proposed to use a Recurrent Neural Network (RNN) in a machine learning method to create a device that can identify bumps and potholes. The device uses an ESP32 GPS Speedometer to locate these bumps and potholes and transmits the position, along with speed, latitude, and longitude values, to a database via Wi-Fi. The device also chooses to save the data it collects in the cloud. With the aid of machine learning, it can be classified into various categories based on requirements. It also has a speech module that allows audio notifications to be played over the speaker. This research suggests using a Recurrent Neural Network (RNN), a machine learning method, to train parameters autonomously in order to enhance vehicle suspension health and prevent driver fatigue.